The goal of this study is to systematically examine the impact of genes on risk of occurrence, age-at-onset, severity, and progression of clinical symptoms of Alzheimer's Disease (AD), and the joint effects of epidemiological and genetic risk factors on these same variables in the white and black populations served by the University of Pittsburgh Alzheimer's Disease Research Center (ADRC). Cases will be classified as familial on the basis of having two first degree relatives with probable or confirmed Alzheimer's disease, or as non-familial. Linkage analysis will be used to classify familial cases as involving chromosome 14, 19, 21 or other unidentified loci and association studies with markers tightly linked to chromosome 14q24.3 and apolipoprotein E genotype (chromosome 19) will be used to achieve the following specific aims: 1. In familial cases, to expand the pedigrees of familial cases and perform linkage analysis to identify families segregating for known susceptibility genes, and to identify potential gene carriers within those families to test the following hypotheses: (1): Inherited susceptibility to Alzheimer's disease is completely explained by mutations in chromosome region 14q24.3, the amyloid precursor protein locus (APP) on human chromosome 21 and/or the apolipoprotein E (apoE) locus on human chromosome 19. (2): Familial Alzheimer's disease in African-Americans is explained by mutations in the same genes identified in Caucasians. (3): Clinical characteristics of patients with mutations predisposing to Alzheimer's disease are independent of the specific locus involved. 2. In non-familial cases, to genotype cases and matched controls drawn from the African-American and Caucasian clients of the ADRC for the apolipoprotein E polymorphism, and for markers tightly linked to the Alzheimer's disease locus on chromosome 14q24.3, to test the following hypotheses. (1): The occurrence of Alzheimer's disease associated with genotypes at the apo E locus is independent of ethnic classification. (2): The occurrence of Alzheimer's disease associated with the apo E locus is independent of epidemiological risk factors. (3): The clinical and histopathological characteristics of Alzheimer's disease are independent of apo E genotype. (4): There is no association between risk of Alzheimer's disease and markers linked to the AIzheimer's gene region on chromosome 14q24.3. 3. In cases of autopsy confirmed AD, to sequence the exons and flanking intron-exon boundaries, and the 5'-promoter regions of the apoE gene from cases and controls homozygous for the apoE E3 and E4 alleles to test the following hypothesis: Hypothesis: The association of AD with the apoE locus is determined by DNA sequence variation in linkage disequilibrium with, but independent of, the substitutions responsible for the well know apolipoprotein E polymorphism. While some of these hypotheses have been tested in Caucasian AD families and cases, they have not been tested in African Americans. The latter is a major focus of this proposal.